Error "can not copy the lexicon" during installation

[gregorw]

Hello TREX Experts,

I have my NetWeaver ABAP 7.52 Developer Edition running in a Ubuntu 16.04.5 LTS using Vagrant with this script: https://github.com/sbcgua/sap-nw-abap-vagrant. From the SAP Service Marketplace I've downloaded TREX 7.10: Revision 77 and I've also tried Revision 76. But in both cases I face the same issue during installation.

As a pre-requisite I've installed libstdc++6 and linked it to the position where the TREX installer expects it. Then I could start the installation:

sudo apt-get install libstdc++6
sudo su -
ln -s /usr/lib/x86_64-linux-gnu/libstdc++.so.6 /usr/lib64/libstdc++.so.6
./install.sh --action=install --sid=TRX --instance=10 --target=/sybase

That is the relevant output when the installation get's stuck:

Installing new version 7.10.76.404644, creating link: TREX_7.10.76.404644 -> /sybase/TRX/exe/nuc/linuxx86_64/trex_new
Copy the lexicon /sybase/TRX/exe/nuc/linuxx86_64/TREX_7.10.76.404644: /sybase/TRX/global/trex/custom/config/lexicon -> /sybase/TRX/exe/nuc/linuxx86_64/TREX_7.10.76.404644/lexicon
can not copy the lexicon


Action:  Configuring SAP system TRX
possess: /sybase/TRX/profile/DEFAULT.PFL
possess: /sybase/TRX/profile/TRX_TRX10_vhcalnplci
*******************************************************************************
Action: Configuring SAP system TRX -- failed. 
*******************************************************************************
############ >>>ERROR:installsapinit.sh FAILED with rc=1280 - TREX setup cannot proceed.
*******************************************************************************
*                   ===  Script execution resumee  ===                        *
* Step 1: Preparation and checks before TREX system installation      : SUCCESS
* Step 2: TREX system installation                                    : FAILED
*
* Script execution terminated at 2019-02-09_09.22.51 with error 1
* See also log file: /tmp/trex_install_2019-02-09_09.22.43/_install.log
*******************************************************************************

This is the logfile output:

======================================================
-----------------  SarTrek.makeitsap  ----------------
------------------------------------------------------
SarTrek.makeitsap
------------------------------------------------------
*******************************************************************************
Action: Configuring SAP system TRX -- failed. 
Exception description:
Traceback (most recent call last):
  File "./support/_install.py", line 1354, in installInstance
    self.memberInstallStatus.doItOrNot(self.memberSarTrek.makeitsap)
  File "/sybase/tx_trex_content/TX_LINUX_X86_64/support/setuplib.py", line 5930, in doItOrNot
    step(*args, **keyArgs)
  File "/sybase/tx_trex_content/TX_LINUX_X86_64/support/setuplib.py", line 2508, in makeitsap
    raise SetupException("installsapinit.sh FAILED with rc=%d" % (int(rc)))
SetupException: >>>ERROR:installsapinit.sh FAILED with rc=1280
*******************************************************************************
############ >>>ERROR:installsapinit.sh FAILED with rc=1280 - TREX setup cannot proceed.
Exception description:
Traceback (most recent call last):
  File "./support/_install.py", line 3566, in main
    Installer().mainTrexAction()
  File "./support/_install.py", line 705, in mainTrexAction
    self.installInstance(step)
  File "./support/_install.py", line 1381, in installInstance
    framework.error(str(e) + " - TREX setup cannot proceed.")
  File "./support/_install.py", line 3476, in error
    raise ErrorException(self.errorMarker + text,result=1)
ErrorException: <__main__.ErrorException instance at 0x7ff4912a1a28>
*******************************************************************************
*                   ===  Script execution resumee  ===                        *
* Step 1: Preparation and checks before TREX system installation      : SUCCESS
* Step 2: TREX system installation                                    : FAILED
* Script execution terminated at 2019-02-09_09.20.13 with error 1
*******************************************************************************

Hope that anyone can help me with that unsupported installation.

CU
Gregor

[Robert_Russell · Accepted solution]

Hi Gregor,

I haven't used TREX in ubuntu but I would suggest the errors point to sapinit. Ubuntu does not have the chkconfig command which prevents sapinit install.

I would suggest commenting out the chkconfig lines in installsapinit.sh script in the install directory "support/installsapinit.sh" or replace/add a line with

systemctl enable ${SAPINIT}

However.

Before installing again, also link sh to bash and not dash as most likely the case in Ubuntu and try again

/bin/sh -> /bin/bash

**even if the above gets past the error - not sure TREX will work on unsupported O/S**

Regards

Robert

[gregorw · Accepted solution]

Thank you Robert, that brought me one step closer to a solution. I was able to finish the installation. At the first start the start failed do to the command lockfile not found. After installing:

sudo apt-get install procmail

The start continues further but now I get the following error in /usr/sap/TRX/TRX10/vhcalnplci/trace/TrexDaemon_vhcalnplci.00000.000.trc:

[139941310355264] 2019-02-11 09:20:37.965 e Daemon TrexDaemon.cpp(01981) : creating nameserver-shm (id=1, size=1073741824) failed with rc=1, os-error=12(Cannot allocate memory)

[139941310355264] 2019-02-11 09:20:37.965 i Daemon TrexDaemon.cpp(06423) : opened shared memory with returncode 0

And /usr/sap/TRX/TRX10/vhcalnplci/trace/startsap_TRX10.log just contained:

protect /home/trxadm/.sapenv.sh against overwriting
protect /home/trxadm/.sapenv.csh against overwriting
updateConfig.py finished at 2019-02-11 09:20:37.917


11.02.2019 09:20:37
ShmDetach
OK
(26873) Starting: local /usr/sap/TRX/TRX10/vhcalnplci/trace/TRX.sapTRX_TRX10 -d -nw -f /usr/sap/TRX/TRX10/vhcalnplci/TREXDaemon.ini pf=/usr/
sap/TRX/SYS/profile/TRX_TRX10_vhcalnplci
(26842) Waiting for Child Processes to terminate.
(26842) **** 2019/02/11 09:20:38 Child 26873 died due to Signal 11 (core dumped). ****
(26873) **** 2019/02/11 09:20:38 No RestartProgram command for program 2  ****
(26842) **** No more Child Processes to wait for.
(26842) Parent Shutdown at 2019/02/11 09:20:38 


Execute Post-Shutdown Commands
------------------------------


11.02.2019 09:20:38
ShmDetach
OK
(26842) Exiting with Return-Code 3. (No more child processes)
 Startup of Instance failed

Perhaps you have some more tips.

[Ganesh119 · Accepted solution]

I have also got the same problem. But for my case, Package "chkconfig-1.20-2.el9.x86_64" is missing on the RHEL server. After installing this package issue has been resolved.